Turquoise Energy Ltd. News #104
  covering September 2016 (posted  October 2nd 2016)
Victoria BC
by Craig Carmichael

www.TurquoiseEnergy.com = www.ElectricCaik.com = www.ElectricHubcap.com = www.ElectricWeel.com

Differential transmission concepts proven by experiments, work progresses (see Month in Brief, Electric Transport)

Month In Brief (Project Summaries)
- UIVT: from Sprint/Variable Transmission project - Conversion of a car alternator to a permanent magnet alternator - Tesla Turbine Windplant Concept

In Passing (Miscellaneous topics, editorial comments & opinionated rants)
- World Nearly Loses an Inventor - Fruit Picking Pole Makes Picking High-Up Fruit Safe & Easy - Tilapia update - The Ultimate Theft: Private Central Banks are Buying Up the Whole World with Money They Print Out of Thin Air  - The Web of Conflicts of Interest of Officials, between the USA Government and "Corporatocracy" - Canadian Parliamentary Voting Reform Committee meeting - Honey Bee Ecological Catastrophe - Improve Health and Increase Life Expectancy: Metformin

- In Depth Project Reports -

Electric Transport - Electric Hubcap Motor Systems
* Electric Hubcap motor, Chevy Sprint & Differential gear Variable Transmission:
  - Sprint Prototype with Third Shaft?
  - Single Variable Pulley with Third Shaft - or Controlled Idler Arm
  - Chain Drive Experiments/Tests
  - Conclusions (proves concepts)
  - Electric Hubcap Motor Troubles (Wiring)... & Electric Caik Motor Upgrade
  - Gear Cutters
  - Gear Making Calculations
  - Other ways to Make Plastic Gears!
  - The Spur Gear Differential Gear Unit & Variants
  - Configuring a Production Model

Other "Green" Electric Equipment Projects (no reports)

Electricity Generation
* Conversion of a car alternator to a permanent magnet alternator with a ring magnet
* Tesla Turbine Wind Turbine?

Electricity Storage - Turquoise Battery Project (NiMn, NiNi, O2-Ni), etc. (no reports)

September in Brief

  Between the experimental work on two projects that turned into about four, news that seemed worthy of note or comment, a close call with a bus, and meetings with Parliament's Voting Reform Committee and a BYD electric vehicle demo, there seemed to be a lot to write about this month.

   A few days were spent in the first half of the month on finishing the conversion of a car alternator into a permanent magnet alternator using a big ring magnet. It seemed to work quite well, but had more magnetic friction than I had expected. (Owing to that friction, near the end of the month I started thinking more favorably of the Hugh Piggott frictionless air core axial flux generators for use in electrical generation projects. The main components of these can be purchased. I started coming up with improvements, especially my rotor magnet strapping system, that might give these units higher power ratings (+50% or better) by getting more cooling air flowing through them.)

   The alternator was interspersed with testing aspects of the new differential variable transmission, proving out the theory, which became the sole focus when the alternator was finished and tested. And that branched out into consideration of how transmission gears might be produced and what types of gears to make.
   I found more ways to make plastic gears in or inspired by youtube videos, figured out what seems like a better way to do the variable ratio belt drive part, and I found a differential gear made entirely of spur gears instead of bevelled gears. It looks better, and it would be easier to make. It's quite similar to my idea for a "double sun" version epicyclic gear, but the casing and construction illustrated was far superior to my vague ideas. It would be thinner left to right than any ordinary differential gear type, making for the smallest - or at least thinnest - "OEM" transmission.
   Along with that work went repairing the Electric Hubcap Motor that was running the Sprint test car. One more time motor and controller troubles got in the way of the transmission project. I found frayed insulation causing a short, and a power wire connection that had never been soldered, that had somehow run reliably for 5 years since making the motor until, I suppose, the short had stressed it.
   But it still wouldn't run right, and to be sure whether it was the motor or the controller, I decided to (at last) repair the Electric Caik Outboard's motor so that I would have a second motor to try out on the Kelly BLDC motor controller. That should have consisted mostly of reassembling it, but if I was repairing it, I should upgrade the magnet attachments so it could do 3000 RPM safely instead of just 2000, which would be much better for the boat. So that project got underway - much belatedly, really, considering the last boat trip and the incident were August(?) 2015. It was the lure of making potentially better motors - unipolar PM, switched reluctance and 'permanent magnet assisted' - that had initially stopped repair, but those projects have now themselves been on hold for some time, without having brought motors and controllers that really perform well so far.

   One day in there an inattentive bus driver ran me, not off the road, but over the centerline into the oncoming traffic lane. I swerved in time not to be crushed by the bus, but if there had been a center divider or any oncoming traffic, #103 might have been the last Turquoise Energy News. I wrote this incident up in In Passing to try and help clear it out of my mind.

   On the 18th I conceived of doing a Tesla Turbine Windplant with a "horn" facing the wind to funnel air into it. This seemed to have the potential for making more power per wind frontage area than any other type, with Tesla's clever invention potentially beating the "Betz limit" by virtue of effectively having multiple turbine stages in one stage, and thus transferring most of the energy and reducing the windspeed to "not much" at the outlet. It would also have no spinning parts on the outside, making it safe for birds and near people, and probably quieter. But my mini Tesla turbine still hadn't arrived at the end of the month.

   I ended the month on the 30th by going to a second meeting about BYD ("Build Your Dream") and getting a ride with the other attendees on the electric bus they had brought to town. On part of the route, an electric Nissan Leaf was right in front of us.
   It seems humorous that I walked to the venue, and at an intersection the electric bus pulled actually up and stopped beside me. I stuck out my thumb and called "Going to BYD?" But it said BYD on the bus and the driver didn't catch on. Maybe if I'd called "Going to 360 Harbour Road?" he'd have realized I must be an attendee and opened the door. But as he turned right and pulled out next to me I really noticed it: No roar! No stench! The bus has a 300Km range and recharges in 3 hours from 480 volts at up to 100 amps.
   BYD makes their own lithium-iron phosphate batteries. Their philosophy is never mind getting the best or ultimate battery specs - just make them as reliable as possible and to last as long as possible. A friend attended who I wanted to exchange the latest with. He suggested coffee, and somehow I forgot to check out the 5 ton electric truck and electric car they had also brought. I also neglected to take any pictures of the event.
   If things go as BYD is planning, they may sell more electric cars in the next 3 or 4 years than everybody else has up to this point. And hopefully quite a lot of buses and some trucks.

A BYD Electric Bus in California

   And I don't know what would stop them from building a similar version with railway wheels for passenger and commuter service, other than having to stop for charging several times per day. Of course strategically placed sections of line with overhead wiring could solve that problem: they could use grid power and charge while travelling.

   After, I went straight to my other destination, Mac's Auto Electric, and bought a 5-Vs poly-V belt pulley that matched the belt I had, for the production prototype of the differential transmission.


Experiments with two chain drives proved the differential drive principle, yielding
forward and reverse and reduction ratios that varied greatly, with only slightly
different size sprockets.

   Having established the principles of an infinitely variable torque converter or transmission, I tried in August to prove them by running the differential drive with simple fixed ratio pulleys, but mostly the belt just slipped, showing that for high torque, very hefty builds are necessary. For September I decided to try the same experiment with a chain and sprockets. Once the problems were (for the most part) worked out this (mostly) worked. I found that 9 to 1 from the motor to the wheel was still a pretty high ratio for getting moving, which surprised me somewhat. It would be a good "second gear" from about 10 to 30 Km/H.
   Then I bought another sprocket that made for both a higher reduction ratio (-16 to 1) and a reverse drive. That is, the motor had to turn backward to move the car forward. Since the motor turns equally well either way this is of little consequence except internally to the gears.
   With the greater reduction ratio the car should have started moving easily on the rough lawn, but with the high tension the chain, not pefectly in line, jammed up. This effect in milder form probably also contributed to the 9 to 1 not moving the car upslope more effectively - smoother should run better.

   The tests did however prove the differential drive theory (in case there was any doubt). If the free left end of the differential wasn't driven, it would turn twice as fast as the center drive and the right car wheel wouldn't turn. Since the differential center drive had a 3 to 1 speed reduction from the motor in its chain drive, a left side reduction of 1.5 (to 1 - twice as fast) would produce "idle" and the car wouldn't move. The lower sprocket I put on the differential right had 36 teeth, so a 24 tooth sprocket at the motor shaft would give "idle". With the 20 tooth sprocket (left end chain ratio 1.80) the ratio between that and the 3:1 drive of the other chain to the center of the differential made for a final drive ratio of 9 turns of the motor for 1 turn of the wheel. The 26 tooth gear (ratio 1.385) gave a final drive of -16 to 1, in other words being less than 1.5 it ran the car in reverse, and being closer to "idle" than the 20 tooth sprocket it gave a higher final reduction ratio.
   Adding the variable drive and not having its belt slip, to make a working "infinitely variable" torque converter or transmission will probably be the largest challenge, but some ideas to help do that have emerged.

   As the month went on, I got gear tooth cutting wheels and started putting together my conception of a production unit. I layed out components (not all the actual ones I would use) on a piece of 11" x 17" paper and started sizing things up.

Laying out components and sizing things up for a production unit.

Motor (& Controller?) Troubles - Caik Motor Rebuild

   Just as I finished the second set of fixed ratio chain experiments, there was motor trouble. I found that the insulation had frayed off the power wires going into the motor and shorted two of them. Then I found a join that had never been soldered, leaving me to wonder why it had worked well for several years.
   When I put the motor back in the car, it still didn't run right, losing power in 1/3 of the positions. Finding no more trouble in the motor, this sounded like blown drivers in the Kelly controller.
   To be sure, I decided to at last repair the Electric Caik Outboard in order to have another motor to try out with the same controller. But I also wanted to upgrade the rotor to having an epoxied strap wrapped around each magnet, which I estimated would make the motors safe up to at least 3000 RPM instead of 2000. This would be much better in every way for the outboard. It required milling a slot in the rotor for each magnet. So the Caik motor rebuild and reassembly became a project in itself. Other matters came up, and at the end of the month the motor still needed installing back in the outboard.

The improved Electric Caik rotor, allowing higher RPM.s
(I had to trim the outer corners off the straps - they rubbed on the outer wall in some places.)

Bulged stator end of the Caik motor.

   As I worked on it, I noticed another reason that the rotor had moved down until it was rubbing on the wall of the stator compartment. The magnetic forces exert a continuous pressure in that direction, and the whole outer wall had bulged out until the rotor touched the inner wall. The upward pushing shaft in the outboard (TE News #90) was merely the straw that broke the camel's back. Whether this deformation was over a long time or mostly right after the moulding I'm not sure. I had originally intended to put the pressed metal bearing holder on the inside to help stiffen it, but there just wasn't room. Maybe some more bolts, between coils and nearer the center, from the outer to the inner wall? Again there's more room in the larger motors.

Conversion of Car Alternator to Permanent Magnet Alternator

   The trouble with car alternators as alternators/generators for windplants and the like is that they have a 'field coil' electromagnet in the armature, which is what the brushes power. This uses power. The more the desired voltage per RPM, the more current has to go into the field coil to give it more magnetism. In one alternator I measured 3 amps at 12 volts, or 36 watts, being consumed by this coil. If your small windplant is making 50 watts in light wind, that would only leave 14 watts output.
   A permanent magnet on the other hand uses no electricity, so the whole 50 watts is available for output. The disadvantage is that its voltage (and the current capacity) rises linearly with the RPM. Typically it won't work at all until the RPM is high enough to give voltage higher than batteries being charged, and above that it will put more and more current into the battery regardless of its state of charge. I decided to put a very powerful ring magnet in in place of the coil. It would probably give more volts per RPM than the coil at full strength.
   There are other ways to regulate output voltage (once rectified to DC) for battery charging or other load. A DC to DC converter with a fairly wide input range will go a long way toward it. And maybe when it's done, the infinitely variable transmission (probably under computer control) can maintain some desired output RPM and hence output voltage level, regardless of propeller (or other input) speed - and regardless of varying loads.

   This sounded like an almost trivial conversion: replace the toroidal coil with a toroidal permanent magnet. But the devil is in the details. I wrote last month of the surprising challenges in getting the alternator apart. Now (September) it had to be modified and put back together.
   There was a 'cylinder' of magnetic material that passed through the coil. That would short out the permanent magnet's  field like a 'keeper'. (At least... I think I have that right. Or I may be wrong and was wasting my time removing it.) So I was going to turn away most of the center iron from inside the coil on the lathe, just leaving a 'button' on each end to mount the magnet on, but owing to various protrusions there didn't seem to be any way to attach the pieces to my small lathe. I had to go down to AGO and get machinist Ralph to do it for me on a very large lathe with a big chuck that could grab it around the outside without bending the fan blades. I had hoped this could be a home project for any handyman. It was certainly not in my thoughts that I'd have to have outside help myself!

Alternator rotor with the electromagnetic coil replaced by a powerful supermagnet ring
magnet. (The spacings between petals were evened out when pressing the assembly back
onto the splined shaft. When free they wanted to snap to one side or the other.)

In Passing
(Miscellaneous topics, editorial comments & opinionated rants)

World Nearly Loses an Inventor

   It was probably quite a rare situation. On the 6th or 7th, I was driving down Quadra Street in Victoria, minding my own business. It's not a street I often travel on. I stopped at a red traffic light. There was one vehicle ahead of me. A bus behind changed into the right lane and came up beside us. The right lane was full of parked cars except for the space the bus pulled into, the intersection, and a bus stop not much more than the length of the bus after it. So I hope the reader will pardon me for assuming the bus must be stopping for passengers, and I gave it no more thought.
   When the light turned green, the car in front of me turned left, and I started moving. The turning car occupied my attention as we all started up. The front of my car was just a few feet behind the bus driver - not much short of being beside him, or should I say under him. As I accelerated, the bus also accelerated. We both went through the intersection, both picking up speed and maintaining the same relative positions. I had expected to almost immediately pull ahead as the bus slowed, but I couldn't. I couldn't believe he had both forgotten I was there and also didn't see me. Where would he think I had gone? What the heck did he think he was doing, with me almost right beside him? By the time I realized that he was accelerating at full throttle and had no intention of stopping, it was manifestly too late to do anything. I could neither pull ahead, nor drop back the entire length of the bus to get behind him. He changed lanes right into the space I was occupying, and I had to steer into the oncoming traffic lane on the left side of the road. Luckily there was no road divider there and no one was coming. I honked and finally the bus driver slammed on his brakes and let me get back on the right side of the road. (Whew, not homicidal after all!) Then, after almost killing me, he had the nerve to honk at me from behind!
   The next traffic light was again red. This time I was the front car. The bus came up from behind again and pointedly changed into the right lane again, and pulled way ahead into the intersection to where I could see a left turn signal flashing in the middle of the bus - which had again hardly finished changing lanes to the right. He pulled the same maneuver. This time of course, I waited until he was well down the street and back in the left lane before I started moving.

   Luckily there were no serious consequences except for the incident haunting my mind, sneaking in unbidden and unwelcome, for weeks whenever I wasn't concentrating on something else... like when I tried to sleep or meditate, and on waking up in the morning. And having to tell myself it was pointless to be paranoid of every other vehicle on the road when driving - especially buses snapped me instantly to an extra adrenalin level for a few days. Being unable to telepathically read the bus driver's intent, I don't know what I could have done differently. If I hadn't managed to veer left fast enough or if there had been a center divider there or oncoming traffic things might have turned out very badly, and if I had had the wherewithal to honk sooner, or managed to gain just a few feet to where he would (at last, surely!) have seen me, the bus might have run headlong into a row of parked cars.
   I kept thinking too of things I might have shouted at the bus driver at that second stop. Given his honk and his further aggressive maneuver at that point, I have the impression that because I didn't respond, he somehow thought the whole thing was my fault. I'm sorry to have left him with that delusion. The more I thought about it, the more I think I should have given him a long return blast from the horn after he honked, or yelled "Watch where you're going!" at that next stop. I hope he gave it some more considered thought afterward, or that a passenger told him I had been there all along. Anyway, I did my best as it happened at the time, as usual not thinking fast enough and being sure enough of myself in real time to give an appropriate response. I refrained from shouting anything hurtful. Finally I decided to write this piece about it in hopes of maybe helping to clear it out of my mind. Anyway, at least I'm still here to think about it.

   Some years ago (maybe it was a decade or two), bus drivers had been complaining that it was hard to get back out into traffic after stopping for passengers, and a bylaw was made that other vehicles had to yield to a bus pulling out from a bus stop. Fine. And I've never had a bus driver expect cars to slam on their brakes and pull out unreasonably from a stop while they are being passed or are about to be. But this bus hadn't stopped or even slowed down. With just two small vehicles ahead (only one at the next intersection), surely he could plainly see there were no passengers at the stop before he changed lanes, and obviously he knew none were getting off. There was, therefore, no reason for his lane changes -- except to take license with the bylaw in order to pass a vehicle or two: on the right, through an intersection, and with nothing but parked cars right ahead if anything didn't go exactly as planned. When I mentioned the incident to someone later, they said "Oh yes, they do that." So I guess it wasn't just this one driver. But if I've ever seen anything like it before, I don't remember it. Someone else said "The taxis and buses seem to think they own the road."

   If any other vehicle, commercial or private, tried to pull such an unusual, unexpected and IMHO dangerous maneuver, the police would be all over them. To do it without even checking his blind spot for traffic for the whole time was very careless. But nobody's perfect. I've certainly made my share of errors of judgement and missed seeing important things while driving over the years. As I say, doubtless the whole thing was a very rare situation, and possibly I'm out of line... but I wish the bus drivers would save their stunt driving for a movie set.
   The fact that my car was metallic dark blue may have been a contributing factor to the driver not noticing me. I prefer bright colors, but it was the best, not to say the only, good deal I found when I was car shopping. The headlights were on, but they were as much below the driver as behind him.

   Near the end of the month a bus pulled a similar stunt in a long lineup of traffic, driving down a street in the "right turn only" lane and then pulling into the through lane in front of all the other waiting vehicles when the light turned green - including another bus which had politely got into the lineup with everyone else.

Fruit Picking Pole Makes Picking High-Up Fruit Safe & Easy

   These have long been known, but they're the sort of product that word doesn't spread about, because no one makes money off it. There are many variations on the theme, but basically one uses a long pole with something on the end to pick the fruit with the picker standing safely on the ground. Mine I made at least 30 years ago with a 12 foot long 1" * 2" board and a large juice can, screwed to one end. A slit is cut into one side. The user raises the the can on the pole and gets the can under the fruit. Then he works the stem of the fruit into the slit and twists. Usually the stem breaks off and the fruit falls into the can. For most apples or pears one can pick two, then lower the can end and release the fruit onto the ground.
   I picked about 8 pounds of apples one day from a tree down by the sea walk, and 9 a couple of days later. The picking didn't take long - it was the carry home that kept me from picking more. My fig tree had a few juicy looking figs way up at the top, too high even for this pole (and invisible from the ground), but it was near the house and I could see and just reach some from an upstairs window. The can being horizontal instead of vertical, 3 of 4 figs fell down to the lawn. My chickens were elsewhere. I had just heard some raccoons go by below and I hoped they wouldn't return. They didn't, but by the time I got down to the lawn, a deer was just munching one down. It backed off a ways and I found the other two. Maybe I should move out to the country where things aren't so busy! (The next time I looked, starlings had found the figs and pecked big chunks out of the riper ones.)

12' pole going up to pick figs. - 9 pounds of otherwise well out of reach apples I bagged.
The slit in the side of the can to break off stems is a key feature.

Tilapia Update

   Well, I didn't get into aquaponics this summer, but I still have the 3 tilapia in a 30 gallon aquarium. Every second or third day I siphon out 2 or 3 gallons of water along with fish poop on the bottom, and replace it with fresh dechlorinated water. It's considerable work, but if I don't do it enough the water gets cloudy and a bacterial bloom starts, and the fish wouldn't last very long. They just keep getting bigger. The 4" pipes for the smaller males to hide in are getting to be too small, and the big female can hardly turn around. But no new brood so far!

The Ultimate Theft: Private Central Banks are Buying Up the Whole World with Money They Print Out of Thin Air

Well, the title pretty much says it all. The US Federal reserve, a private corporation as are all central banks in the western world, along with the other big banks, collectively own outright over half of all real estate (as I understand it, by property values) in the USA. They are also the largest landlord. After they [claimed they] stopped printing 85,000,000,000 $/month, so-called "quantitative easing", the BOJ (Bank of Japan) went on a printing spree, and then the ECB (European Central Bank) started printing up 80,000,000,000 Euros/month. Presently global printing is said to be at a rate of 160 billion $/month. Next will come "QE4" from the "Fed", who are speaking of printing another 4,000,000,000,000 $. That four trillion will just be for starters. If the printing stops, the financial and credit system will immediately crash, which will then crash the economy. This has now been going on for nearly a decade, multiplying the amount of currency previously in existence by many times over. The only thing that's prevented hyperinflation so far is that the "money velocity" is at historically low levels, that is, that most of that money isn't really circulating through the economy - yet.
   However, the central banks have been buying up government bonds ad nauseum, and now governments haven't been taking on more debt fast enough, so they've started buying up corporate bonds and company shares - thus also boosting the bond market and stock market bubbles so they don't come rapidly crashing down. One result is they are also the major players, probably owning large percentages of the total outstanding stocks in every stock market. And they give loans at almost zero percent to large corporations, which instead of investing in growth use the money to buy back their own shares while they lay off more workers. This again boosts the stock market, and makes the 'earnings per share' stats look better. It's all manipulated to boost published economic figures - while jobs disappear and more and more of the whole population is marginalized. The big banks that made far too many bad loans, for too low rates, highly inflating asset prices in the process, will all be bankrupt if those inflated prices are ever allowed to correct to fair value - to where citizens can afford houses et al again.
   Another result is that the companies taking advantage of these offers are so highly leveraged that when business turns down for a bit, they become bankrupt. Hanjin shipping (7th largest in the world?) went out of business in early September leaving 14 billion dollars worth of cargo floating abandoned in the world's oceans. Their assets were frozen and they couldn't even pay the port fees to land the cargo - goods purchased by the stores for next Christmas season. The abandoned crews are running out of food and jumping ship, leaving them as floating ghost ships. Don't the banks even want to recover the ships and cargoes, or retain the crews? Maybe that's too much trouble - just get more free money instead? The banks have been less than good managers of the assets they gobble up. Hanjin is only one of many companies now deeply leveraged which will probably go belly up in the near future. (This also demonstrates how the entire distribution system will freeze up when credit dries up.)
   So the banking system keeps itself in business by buying up everything so prices don't drop, with payments to politicians to keep governments in collusion and have special laws written for them, while the real economy sinks under the burdens of inflating prices with (on average) dropping incomes, and the artificially created scarcity.

   The Swiss and Norwegian central banks recently (~Sept.8) printed a billion dollars worth of currency and used the money to buy up shares in mining companies - a clandestine means of buying gold and silver without crashing the COMEX and LBMA and driving precious metal prices way up - which might by itself initiate the economic implosion.
   It's all theft on the largest scale. Those who print the money are becoming the owners of every tangible asset on the planet, while the whole population is being gradually stripped of whatever they have.

   Most people, using their increasingly hard earned money to buy things they need, are now in direct competition with those who freely print the money, whenever they try to buy any form of real property. Are they not being cheated? Are the 99% not getting shafted? On the 22nd it was reported that almost 100,000 people in New York had applied for one-time emergency assistance to pay their rent. All over the USA, people are living in campers, vans or even cars, or are entirely homeless. Where are the core values of social sustainability - Quality of Life, Growth Conditions and Equality? They are going down the drain in a sea of debt owed to those few who so freely create the money. Is this not the world's largest bubble in all of history, with virtually everyone heavily in debt (owed to who?), which must soon pop?

An aside: Growth of societies that suddenly collapse with huge losses of population has been the history of this planet so far. They are usually preceded by unsustainable population expansion. You've heard of the fall of Rome. But did you know it was preceded by such onerous taxation over the decades that many peasants abandoned or lost their land and became slaves to large villas, or went to live with the barbarians? With such inequality there were repeated bouts of plague, and no thriving rural population in the Italian countryside to draw from when Rome needed soldiers to defend the city as there had been in the times of the republic. And you've heard about the Black Plague that killed over 1/2 the population of Europe from around 1350 to 1450. But have you heard about the banking collapse of the 1340s that cut off all credit, impoverished Europe and so paved the plague's way? And you've heard about the Irish potato famine, but did you know it was the introduction of the potato to Europe from the new world that had first brought about a rapid, unsustainable growth of the Irish population?

   The juggernaut forces hurtling us towards the economic cliff are almost global, or at least they pervade the entire western financio-economic-political system, as David Quintieri so clearly illustrates (see next item). Climate disasters, huge die-offs of life (mostly caused by geoengineering?) and (for whatever reason) a great rise in occurrence of physical catastrophes everywhere threaten food supplies at their source. Huge population bubbles in eastern Asia and Africa, along with antibiotic immune disease organisms, are setting things up for a sweeping plague or plagues. The crash or even a series of crashes can hardly be prevented or too much longer delayed at this point.
   We can however, educate ourselves to understand what is happening and why, and perhaps begin to formulate and discuss solutions based on the core values so that this will be the last time it happens on this planet. After some real, population crashing disasters both environmental and socioeconomic, now complacent people who survive will develop enquiring minds, ready and receptive to new solutions: environmentally sustainable technologies and socially sustainable institutions. eager for them. Will the new solutions be available for them, or must we enter a new dark age? It's always up to individuals, and everyone can play some large, medium or seemingly small part in making the world better instead of worse - improving the quality of life, growth potentials and equality among all.

The Web of Conflicts of Interest between the USA Government and "Corporatocracy"

   One sees that the USA government for decades now has always put corporate interests ahead of the public interest. Perhaps most visibly Washington defames peaceful leaders to invent enemies and puts war ahead of peace on behalf of the arms merchants. They are nearly always at war.

   David Quintieri has done a youtube video that clearly shows why this is inevitable: in a series of Venn diagrams he gives names of an amazing list of people who occupy both a high, influential post in government (eg, congressman, senator, senior civil service, committee head, regulatory agency head...) and a high, influential post in a huge, corrupt corporation (eg, CEO, CFO, board member, head lawyer...). There are separate diagrams for Big Banking, Monsanto, Big Pharma, General Electric, big military industrial corporations, and some others. You will see some perhaps familiar names.
   These rampant conflicts of interest show how the whole lot of them are in it for selfish gain and power, and why they will never as a group do anything to try to fix the problems. (One always holds out hope for individuals, but any there may be are bucking a flood tide pouring over and bursting through all the levys.)
   Sometimes they are even conflicted within themselves, such as when the Pentagon objects to the Navy developing a new weapon (magnetic rail guns) instead of buying existing munitions from the ensconced arms merchants, and when CIA (cocaine importing agency) backed rebels in Syria fight with Pentagon backed rebels.

Title:  Give Me 7 Minutes and I'll Prove Google and Government are One Entity (And More)
By: David Quintieri
Date: August 15th 2016.

   As I indicate, this little viewed video goes far beyond Google and constitutes the most convincing exposé of the collusion and connections between government and parasitic huge corporations I've ever seen. I think it should be given the widest possible distribution!

   Dr. Paul Craig Roberts (co-father of "Reaganomics") also speaks of "the deep state" and why it won't matter much who is elected president unless he can immediately replace a whole host of officials. "We faced this with the Reagan administration. It was a real knock-down, drag-out fight to get anybody on the administration who was on the president's side." says Roberts. "When he appointed me, they put a hold on my appointment. They didn't want anybody there who was going to do what the president wanted." He says a study has shown the public today has zero influence over decisions made by government. It all makes perfect sense once you've watched Quintieri's video.

"When plunder becomes a way of life for a group of men in a society, they create for themselves over the course of time a legal system that authorizes it and a moral code that glorifies it." -- Frédérick Bastiat, 1848

Voting Reform Committee Meeting

   I have previously mentioned the ideas I've written about in http://www.HandsOnDemocracy.org .

   Now the Canadian government, having promised in the last election, has set up a 'voting reform committee', and it - about a dozen members of parliament (MPs) - had a public meeting on Tuesday September 27th in Victoria, which I attended along with 100 other people. There have been similar meetings in other Canadian cities.
   Someone from the 2003 "BC Citizens Committee on Electoral Reform" was featured and the MPs asked him a lot of questions, which he answered very well and articulately. Perhaps all their work was not in vain after all! I hope the sessions in other cities went half as well as the one here in Victoria.
   Members of the public had 2 minutes to speak. I mentioned something about "common ideals, the core values of quality of life, opportunities to grow into one's potential, and equality", and the HandsOnDemocracy.org URL, and said that the subjects covered at the meeting were a subset of a larger puzzle. I gave out a few cards I printed with the URL, including to a few of the MPs.

   I don't think there has ever been anything like this before, and I'm glad I went. It's along the lines of the national government becoming "a learning organization", able to evolve from the petrified forms of yester-year toward more advanced procedures, with a "design team" of a dozen members charged with fact finding, public consultation and formulation of whatever they formulate. It's a very promising development, and perhaps historic! I should have stayed to the end, but it was past suppertime and I left a while after I spoke, with another 50 people (nearly 2 hours) to go.

   Of course not all ideas from everyone agreed entirely with mine, or with each other, but assuming they are actually listening (it seemed they were) - and if the government elected by the present system is actually willing to change it - there will be changes made to the way Canada votes, and they will be improvements. I was pleased to hear several people speak who recognized how the present single ballot "X" voting system has polarized politics, and some other sentiments and ideas on voting and elections similar to some on my web page.

Honey Bee Ecological Catastrophe

   In fear of mosquitos with zika virus (an obscure African virus which is (or until recently was) available for sale in vials on a Rockefeller website for about 600$), South Carolina sprayed a nicotinoid pesticide (one that's banned in Europe) to kill them. They got an environmental disaster as it killed every honey bee in at least one county. Entrances to hives everywhere were littered with dead bees, a symptom of acute pesticide poisoning. Doubtless other species of bees were also killed off. We may not hear more on this story, but the area will surely be impoverished in life next spring and perhaps for several or even many years, as honey bees are the most important of all plant pollinator species, and animal life depends on plant life. Depending what's grown there, the area extent of the die-off, and how fast bees move in from elsewhere (assuming the pesticide dissipates over the winter), they may get no crops next summer.

(Weeks later:) The spraying continues in the region. It's just one more in a long list of ecological disasters, which are mostly man made. On October first, Full Spectrum Survival youtube channel's Daily News Update reported that a number of USA bee species have just been placed on the endangered species list.

Improve Health and Increase Life Expectancy: Metformin

   Life expectancy has been on the rise for quite a while. When one looks back at the earlier music composers from the 1400s to the 1700s, one sees from their birth and death dates that many (eg Mozart) died in their 30s. In the last couple of centuries people often made it to 80 or 90 if nothing struck them down. Edgar Casey was occasionally asked about "How long should I expect to live?", and the answer was usually "around 120 years" - for one person, "140". I also hear that on very advanced worlds the people can live for hundreds of Earth years.

   Why is it virtually no one makes it to 120 or 140 or beyond? What are we missing, or how are we harming ourselves, that we never get there? We don't know. But there is a drug that holds considerable promise for changing that. Metformin was created in 1922, and started to be used as a type 2 diabetes medicine. Once it had that label, any idea of using it for other purposes was overlooked or forgotten. But in recent decades it's been discovered that it has a whole host of health giving properties. It's anti-cancer, anti-obesity, anti-diabetes, anti-Parkinsons, anti-dementia/alzheimers, probably some other things I don't remember -- and anti-aging. A real miracle drug!
   It has mild side effects, which may include: gas, upset stomach, 'loose stool' or even diarrhea. These usually subside if it's being taken regularly. (I can get those from eating chilli, which is non-prescription.)
   It has received glowing reports from all the more recent research as a means for greatly extending one's potential life span. The Articles mention the same "120 years old" figure, or "live 40 years longer". And not only will it help one live longer, but those years will be healthier.
   I am somewhat dubious of the "40 years more" claim. When people find out something new, they usually seem to overstate the case by a good margin. Still, if it's 5 or 10 extra years, and if those are healthier years, is it not worth taking? But if "40 years" isn't here yet, perhaps with further research still better things will be created.
   The other technique of course is for people to consciously select partners and put an emphasis on selective breeding to gradually create better and longer lived races of people with fewer and fewer genetic defects. We know more about genetics now, but even in ancient times people selectively bred plants and animals with good results. And once steps are being taken to limit the global population to prevent yet another repeat of the "grow until collapse" scenario now playing out, it will be much easier to say, "you guys are great, go ahead and have 4 kids", or "could you please limit the size of your family, or not have one, in the community interest?" Such measures would of course have to be by general and genuine social agreement, not imposed from "the top" down. Now I have seriously digressed! Working from both angles, I'm sure such age figures will someday be attained and then surpassed by most of the population, who will then view our shorter lives as a part of history, the "bad old days".

   My sister in law has diabetes and takes four 500mg tablets of Metformin a day - 2000mg. I believe it's what got her (and another diabetic I know who takes 1500mg/day) off the insulin needle. Her doctor has said "They should put it in the drinking water!", which he probably wouldn't actually recommend if pressed, but it's a powerful endorsement. No doubt he takes it himself.
   Again proponents are likely to think that if some is good, more is better, overstate the case and suggest taking more than is really helpful. (One doctor thinks everyone over 40 should take 2mg of melatonin per night as "the body no longer produces enough." The pills come in 3mg and 10mg sizes. I find at age 61 that 1mg really helps me stay asleep at night.)
   Perhaps 250mg of metaformin a day would be a good dose for healthy people. 500mg would surely be the top. Side effects would doubtless be minimal if not unnoticeable at low doses. And maybe it would be good for people over 50 or so -- or younger if there are any signs of troubles it can prevent. And potentially it's cheap - "16¢ a pill". But having been prescribed for diabetes, it's become a prescription medication and you can't buy it unless your doctor says so, and apparently unless you have diabetes most doctors won't prescribe it. And even if you get a prescription, it will cost a lot more than it would if it were freely stocked in store shelves.

   Considering all its healthful properties for so many things and its few and mild side effects, this is a pointless and sorry state of affairs. It's not a dangerous substance that needs to be carefully administered, and it's not an antibiotic whose use would be controlled to prevent pathogens from becoming immune to it. It should be freely available over the counter to anyone who wants it. The very fact that it is prescription doubtless prevents people from hearing more about it. Tylenol overdose causes 50000 deaths a year in the USA and that's considered acceptable. Metformin will doubtless be hardly a blip on the radar screen, if that, by comparison. It would become matter of course for the doctor and pharmacist to mention it if something is prescribed that doesn't mix well with Metformin.
   If Metformin is anti-cancer et al and extends health and life, there's a good argument to be made that keeping it "prescription only" is in effect killing millions upon millions of people, making them less healthy and ending their lives needlessly early.

Nepot & Sons

   Far and away the best example of nepotism is "royalty". Take a nation's top political post (head of state), get your entire family in on it from grandma to the kids, and pass it on to your next of kin in perpetuity.

An astronaut? What's that? Man: "What do you want to be when you grow up, son?" Kid: "I want to be a juggernaut!"

Newsletters Index/Highlights: http://www.TurquoiseEnergy.com/news/index.html

Construction Manuals and information:

- Electric Hubcap Family Motors - Turquoise Motor Controllers
- Preliminary Ni-Mn, Ni-Ni Battery Making book

Products Catalog

(Will accept BITCOIN digital currency)

...all at:  http://www.TurquoiseEnergy.com/
(orders: e-mail craig@saers.com)

Electric Hubcap Motor Systems - Electric Transport

Electric Hubcap motor, Chevy Sprint & Infinitely Variable Transmission
...& Electric Caik motor

   A thought occurred to me on the 4th. If the left end of the differential is speeded up instead of slowed down, in addition to reversing the direction the torque required of the variable pulley drive should drop. So, decreasing torque from it as the car heads towards highway speeds. The difference between say 3 to 1 and 9 to 1 is substantial. However, that's backwards - it doesn't help much for getting it going. with 3x reduction on the fixed side, the neutral point is 1.5x either way. The difference between 1.3 to 1 and 1.7 to 1 for a "low gear" isn't a lot. Either way it's a lot of torque. Just a thought.

Sprint Prototype with a Third Shaft

   In August a fixed ratio V-belt drive experiment failed to perform in place of the planned variable drive because the belt just slipped under load. Owing to the high differential ratios, it didn't need to slip much to prevent the car from moving. (I should have tried sanding the smooth enamel painted pulleys to roughen them up to grip the belt better.) I started to think it would be necessary to make something like a production transmission unit, placed before a final drive reduction, in order to lower the torque loads to the variable section, before the Sprint could be made to run the way I intended.
   Then I conceived that if a third, intermediate, shaft was added, the unit could have a lighter variable section running at higher speed to that shaft, then a chain drive reduction to the differential gear. This would lower the torque requirements and also add flexibility, because this new shaft could be placed so that any reasonable length belt in the variable section could be the right length.

   On the 3rd I made another shopping trip to Princess Auto. I bought the asymmetric "CVT" belt, a 36 tooth sprocket gear and hub for the end of the differential, a 16 tooth one to connect the 3rd shaft to that, and a 20 tooth sprocket and hub for an experiment. These were all for #40 chain which I have some of, and I got a couple of links and half links for it to close the ends.
   I also looked at, but didn't buy, one of the complete "CVT" units. Both ends were unsuitable. The input pulley was operated by centrifugal force, which made it operate backward to the desired configuration. The output pulley had a centrifugal clutch, which only went to a small, lightweight chain sprocket, maybe #30 chain, 12 teeth. There was no way to connect it straight to a shaft. And the input pulley was smaller than the output, which (see below) was also backward. But if nothing else, I learned that the asymmetric drive pulley angles are 18° and 2.5° - not quite straight. 2.5° makes sense as the belt might rub on the outside of the pulley if it was absolutely right angle.

   The 16 tooth to 36 tooth chain drive gives a 2.25 to 1 reduction, so the "CVT" will run that much faster with that much less torque load than if it went straight to the differential. (still a pretty hefty load!) So the lowest desired reduction would be 1.333 instead of 3, and the idle point .667 instead of 1.5. In other words, the speed is actually increased for 'low gears' and the output pulley need be only a little larger than the input one for the 'high gears'.

Single Variable Pulley with Third Shaft - or Controlled Idler Arm

   Usually with a "CVT" unit, if one pulley expands the other must contract in order that the belt length remain correct. And this also allows a considerable variation in reduction ratios, over 3 to 1 or maybe 4. On the 18th I conceived that with only needing a 2 to 1 change in ratios to get a huge difference in final drive ratios, only one variable pulley would be needed. The other end (either end, as convenient) could have a fixed size pulley. But then of course, the distance between pulleys would vary. But in fact, that could be a big advantage: changing the distance between the two pulleys, one being a spring loaded variable width pulley, could be the means to vary the 'gear' ratio.
   The third shaft could pivot on an arm with a pivot bearing at the differential. No matter where it was pivoted to, the chain length between the third shaft (now the control shaft) and the differential would stay the same. Thus the chain drive would be unaffected by the pivoting. The third shaft could move toward and away from the motor shaft in a small arc, forcing the spring loaded variable pulley to expand or allowing it to contract. No connection to a variable pulley to move it in and out would be required - moving the arm would do it. That of course would be accomplished via the cable to the shift lever in the car. And maybe later under completely automatic control.
   So, it won't need a throw-out bearing and "clutch forks" to move one side of a pulley along the shaft after all! It's still not going to be trivial to put together either the prototype or a "production" unit, but each new idea so far seems to be making it simpler.

   The spring loaded variable pulley I got goes from about 2.5" to 4.5" effective belt diameter, a variation of 1.8 to 1. Since the motor has its own reverse, that should be good enough, with the right size fixed pulley. The symmetric belts seem to fit well right at the outside of the cast steel regular V-belt pulleys, tho cast aluminum pulleys seem too narrow. If anything, they seem less inclined to slip than a regular V-belt.

Next Experiments: Fixed Ratio Chains

   I thought that the fixed differential drive ratio experiment should be repeated, with a chain drive instead of the V-belt. The chain drive couldn't slip. (unless a sprocket gear slipped on the shaft - which did happen with the belt drive. It's a lot of torque!) So it should move the car this time, unless some other problem introduced itself.
   For this I got the 20 tooth sprocket, providing 1.8 to 1 reduction to the 36 tooth one, which would be installed on the differential left end both for this and for the final configuration. Since it'll be going direct to the differential for the experiment (no 3rd shaft yet), that's turning somewhat slower than the 1.5 to 1 idle point. A 24 tooth sprocket would have given the idle point ratio, where the car wheel wouldn't move at all. I had wanted a 21 or 22 tooth to be fairly close to that but not on it, but there were no sprockets available between 20 and 24. There was a 26 tooth sprocket that was closer, on the other side of the idle point. (Who cares which direction the motor must run?) Why I didn't buy that one instead? I don't remember. If the 20 tooth was geared too high for getting moving I'd go back for the 26.

   I bored out the 36 tooth's hub to 30mm (from 1-1/8") for the shaft that fit into the splined differential's socket, then welded the hubs to the sprockets. I just tacked on the 20 tooth one so I can grind it off later and re-use the hub if I want a different size. In spite of buying one of them new-fangled auto-darkening welding helmets, my welds with the new MIG welder look just as professional as those done when I first started with the big stick welder a decade ago. Not that I get much practice.

4th: Starting before breakfast, I fitted them on and turned the motor by hand. The car seemed to want to move but didn't. I drew some felt pen marks on the shafts. Sure enough, tightly as I'd done up the set screws, the sprocket at the differential was slipping. I took everything apart. I tried to mill a key slot in the lower shaft, but it just dulled the 1/4" HSS end mill, badly. And somehow that turned out to be the only 1/4" mill I had. I gouged a rough keyway into the shaft with the angle grinder and ground a key down to fit it. Then I resharpened the mill (not very well) and did a keyway in the upper shaft before the upper sprocket had a chance to give similar trouble. I made a key for the slot, and then spent some more time putting everything back together and into adjustment again. Sometime in that I had breakfast after 2PM.
   I powered up the motor controller and got in the car. It didn't seem to want to move backward, but it moved forward a bit, then when I flipped it to 'back' again, I had rocked it out of its little pothole and it went - for a couple of feet. Then the chain started skipping.
   A bolt in a slot, which is done up with the chain tight (or now, with one tight and one almost tight) had slipped. Tight as I did it up, when the going got tough it kept slipping and slacking off the chains, which then skipped. I already knew it needed some way to positively lock it into position. I did it up one last time, pressed on the pedal gingerly and drove a few feet into the shed to park. The day was pretty much gone and I had other things to do.

5th: I found where I could put in a thin wedge of steel and put a nut on a bolt to hold it in place, to hold the chain tension adjustment from slipping. It was a bit makeshift, but the simplest solution and it worked. When the third shaft is added things might need changing anyway. I backed up the car a few feet and suddenly the motor was spinning freely. Sigh! The thin washer with a slot in one side to slip it in, to hold the shaft in the differential, had bent and it had loosened off of the splines that lock them together. I had been worried about that weak spot. I found a fatter washer and cut a slot in it, but I had to grind a little off the head of the bolt and then mill down the path of the head along the slot in order to force it into the thin space. Tightening the bolt seemed to pull the shaft farther into the differential socket instead of bending the washer, so it was better than the first one. I'm still worried the bolt will unscrew itself while driving.
   Then I drove the car back and forth a bit. Again it didn't start moving readily, and the slowly turning motor got hot pretty fast. It probably would have been a pretty good "second gear" ratio from say 7 to 30 Km/H. When I went between forward and reverse there was a lot of slack after the motor started turning before the car jerked into action. The chains were both pretty tight, but there's a lot of play in the differential gear. The play is magnified by the reduction ratio.

   Recalling that 24 teeth would have been "idle", using the 20 tooth gear had not made for a really low final differential reduction ratio, 9 to 1, and the car didn't start moving readily if it was uphill. A larger motor should have had no trouble even on hills with this ratio, but it's part of the plan to run a car with a smaller motor, and with a variable drive ratio one should be able to get close to the 'idle' ratio and run at high reductions of 15, 30 or even 50 to 1. (Obviously the final drive ratio obtained can be calculated by some simple pretty formula. So far my mind has rebelled at attempting to figure out what that formula is.)

   Nevertheless, I found out later when I tried a 26 tooth sprocket that things were jamming up when the torque needed got higher, ie, when going uphill. Without that unsuspected problem, the 9 to 1 would surely have worked better. On to that...

   The 26 tooth gear would be 1.38 to 1 against the 36, not a big speed change from 1.8 in itself, but it provided about -16 to 1 final ratio from motor to wheels. That should have been nearly double the start-up torque - also reversal of motor versus car direction. Since the motor runs either direction fine, direction isn't important except it would be nice considering wear for the differential gear to slow down as the vehicle speeds up, rather than speeding up. OTOH, in a production model, the 'differential' would be one that was made for the job, to spin freely in air without issues, so that should be no concern except for this prototype. A 21 tooth sprocket - the size I would have got if they'd had one - should have been about 13 to 1. At this point I wanted to try the 26 tooth sprocket, but it was labour day and the store would have been closed, so that was it for the day.

    The next day I bought the 26 tooth sprocket and a hub. But someone had put a 7/8" bore hub in the 1" bore bin, and I neglected to check when I bought it. So I had to go back the next morning (7th) and get the right hub. I welded the sprocket to the hub first thing, because it looked like it might rain and I couldn't weld if it was damp out. Then I got out another piece of chain and made one to fit.
    Basically everything checked out according to theory: forward became reverse, and the motor turned about 16 times for one turn of the wheel - a very high reduction ratio. Or it would have been if it could go a whole turn of the wheel. One would expect the car would move readily, uphill and down, with this high reduction ratio. But it didn't go far before things seized up. Somewhere, when the torque got too high, something was jamming. With this sprocket, one chain is highly taut at the front and loose at the back, while the other is opposite. Finally I had to slack off the chains to get the seized-up differential to move and the car to roll again, at all, in either direction. It was after this that I realized that high friction was doubtless the same problem the 20 tooth ratio had had with going uphill. It wasn't that 9 to 1 was necessarily too small a reduction ratio for small hills. This 26 tooth one, giving 18 to 1, actually worked much worse!
   I suppose the tension on the chains, and hence the propensity to develop high friction, is multiplied by the reduction ratio? That doesn't sound right... Could there be some hidden, inherent problem in the whole idea of this torque converter? The cause needed to be identified before anything else.

The 26 pin and 20 pin sprocket sizes compared.
This illustrates the effective amount a variable pulley would have to change to go between
"reverse" and "2nd gear" ranges. And highway speeds would be somewhat smaller yet.
With a motor that reverses electrically, it wouldn't need to expand much beyond the smaller size.

   I would have put the 20 tooth sprocket back in and checked for friction - turning the motor by hand reveals it just as well as running it - but it started to rain. The next morning (8th, sunshine!) I went out and looked again. This time I stood inside the hood with one foot on the end of the shaft sticking out of the differential, and started turning the motor by hand. And this time it didn't jam up until it hit quite a steep uphill point, when the chain no doubt had more tension on it than my weight. I thought that was it! The differential gear could handle heavy symmetrical loads, but with only one bearing on one end of the shaft, the left end couldn't handle sideways pulling loads like a chain or belt. And with the center gears running around the middle through opposite points, it may have worked (eg) at 0 degrees and 180, but not at 90 and 270°. That might explain some of the effects, like with the 20 tooth starting up okay and then stalling - it wasn't actually hitting hills, it was reaching a bad point of rotation. I was glad I had left the long end on the shaft so that I was able to perform this test. Otherwise I might have cast around looking for other causes for some time. But the next day I did find another cause.

Second Fixed Ratio Chain Tests

   At first I despaired of finding some way to mount a bearing to take all the force on the outside of the shaft. It would at least involve taking everything apart and welding something new onto the case. Maybe I should be thinking instead of a production prototype after all? Then I thought that I could simply make a bar with a bearing at each end and put it between the two shafts. That should exactly counter the force of the chain between the two, and luckily I had made the top shaft extra long, too. That should do for experimental purposes, anyway.

   On the 9th I made a 12" x 4" x .25" aluminum "bar" piece to hold two needle bearings (luckily I had them) at 8-7/16" apart, the approximate distance between shaft centers with the lower one between flopped down and pulled up, and with the main chain relatively tight (fixing the position of the upper shaft). I cut out the bearing holes with a jigsaw. Filing them out until they fit took quite a while, starting with the cuts a little undersize in many places. If the holes were too big they would fall through and my mounting scheme wouldn't work.
   The needle bearings, each with a stamped-out bearing holder pressing them into their holes in the aluminum piece, can lock to the shafts with a setscrew - vital since there were no attachment points except the two shafts. I wasn't sure 8-3/8" wouldn't be better, but it could be adjusted slightly closer if desired by sliding the top bearing along the shaft before doing it up, with the bar angled a bit between the shafts instead of straight on. I thought this arrangement would be adequate to prevent the differential gear from jamming.

   The result was disappointing. It seemed to jam almost as badly. I finally realized that the two shafts didn't quite line up. The chain fit a little differently than with the 20 tooth and pulled the top shaft down more, twisting it relative to the lower one. And then I realized it was the chains, just slightly out of line, that jammed at high torque as each new link, under high tension, was being pulled onto the driving shaft's leading sprocket. This was unexpected to me. Of course, although it had been quite a while since I had made the original chain drive, I had never tried designing something with a chain drive before. Working on bicycles as a kid... the bikes were already made, lined up! And the torque was much lower and the chains longer, reducing the effects of the misalignments of 10 speed gears. I hadn't realized that the sprockets have to be in absolutely precision alignment, at least if there's a heavy pull.
   This wasn't a problem (AFAIK) with the original chain. (...or was it part of the reason the earlier converters only seemed to work up to a certain torque level? Hmm!) But the new one, being badly out of line with the 26 tooth sprocket, worked very badly. (Maybe the chain needed another 1/2 link?)

   I put the 20 tooth sprocket and chain back on, but it didn't seem up to pulling the car uphill. In fact, I found I couldn't push it up either, nor crank it up with the torque wrench on the wheel, which pinned out at 150 foot-pounds or more. That was a first! It had never needed that much torque at the wheels to move it before. That upped my previous estimate that 150 foot-pounds would be enough to get the car moving under any normal circumstances. It needs to have 200 available. Of course it will get it if the variable converter is working right.
   I had rolled - driven - the car back farther than usual. The back wheels were definitely going upslope. I decided a front one must have fallen into a pit the chickens had dug, and in fact I had to jack one up and put a board under it before I could finally rock the car ahead. (I'm not sure about the chicken theory, but all 4 wheels had to get out of their ruts to move it.)

   Well, misaligned chains and shafts were a big problem in the prototype. Of course in a production unit, everything would be perfectly aligned by design, not thrown together. One expects perfectly aligned chains and sprockets wouldn't jam... would they? Was it perhaps time to give up on the persnickety prototype and start designing a production one after all? Let's see... if I made flat plate sides, I could have them cut from aluminum or steel at a CNC waterjet place. That would avoid making cast pot metal cases and still effect precision placement of mounting points to avoid potential misalignments.

Chain Tests: Conclusions

   The experimental tests were a success, and proved to be valuable if not essential to the general progress. It shows the value of doing such an experiment even tho the results would seem to be a foregone conclusion.

   As expected the reduction ratios between motor and wheels were demonstrated to be based on the difference in ratios between the two drives to the two components of the differential gear, and as that ratio approached 2 to 1 (or in this case 3 to 1.5), very high overall reduction ratios, of 9 and -16 to one, were attained from a single stage of reduction gearing. Relative to motor direction, the car moved in one direction below [3 to] 1.5 and the other direction above 1.5. It was pretty obvious that it would have to, from a description of the operation or a drawing, but here was proof that there were no flaws in the logic.

   They also demonstrated once again that the forces to move a car are very substantial (in case we didn't already know that) and that unless everything is robustly and properly built, something will slip, give out or jam before it obliges by moving. That the poorly aligned chains, and perhaps the unsupported left end of the differential, were jamming - the unsuspected cause of frustrating problems - was disclosed. If I had built the variable belt drive prototype without doing these experiments, the belt would surely have slipped because of a jamming chain, and without understanding the real reason, the whole project might have been once more set aside, perhaps indefinitely, in frustration.

   And not all the effort was single-use: The 36 tooth sprocket on the differential left end was to be the finished component.

Motor Trouble & Electric Caik Motor Upgrade

   After I got the car out of the potholes I ran it back and forth a couple of times, noting that it went okay except a couple of times it either jammed up or couldn't make an uphill bit - I'm not sure which, as they would both happen at similar higher torque points. Then, at the end of the day and having done all the tests, the motor ceased to run. In the evening I checked out the repeatedly blinking codes on the red light in the controller: [... ..   ... .  ... ..   ... .] Um, "reset" and "frequent reset". "High current or momentarily low battery voltage." Could the batteries be low already? I charged them overnight and in the morning (10th) I checked the connections and put a meter on the power at the controller. Everything seemed fine except the blinking light on the controller. Then I wiggled the wires at the motor. The controller then stayed green, but the motor wouldn't turn. Okay, that would indicate: One of the power wires had come loose, and it was shorting to another. When I wiggled the wires, the short came disconnected, but the motor only got power in two positions out of six, when just the other two wires were powered.
   When I opened the stator of the motor (12th), I found frayed insulation just where the wires entered. There was nothing disconnected. That seemed very bad, because when I had un-shorted the wires in the car, presumably it should have run. The fact that it didn't and yet had no bad connections now seemed to point to trouble in the motor controller, no doubt caused by the short. It was some time before I ventured to continue working on it. The wires had cloth insulation (old salvaged wire from the days before plastic (?) coated wires.) and I had put motor sleeving over them. All had worn through with vibration - physical and probably electromagnetic. My opinion of wire sleeving as insulation went down drastically. Apparently I needed to improve the arrangements! The first thing that occurred to me was that most motors have a wiring box on the outside. The second thing was immobilization, which would presumably be with epoxy for this motor. The third was that since my motors have plastic bodies, I could easily route the 3 power wires through 3 nearby but separate holes, where none could touch each other. I picked that solution.

But what about the fact that it hadn't run when un-shorted? It seemed worth going through it and checking for any bad connections. But where might a bad connection be? Maybe it was just some anomaly? But it was the 17th before I closed up the motor, and by then I had put the problems out of my mind. It took it out to the car, and it didn't run. Finally I got a multimeter and checked. There seemed to be an open circuit on one of the power wires. I didn't entirely trust the meter's test leeds and tried several times. Nope, there seemed to be an open circuit.
   I took it back in and opened it up again. There didn't seem to be anything to be seen. I turned off the lone marrette connector that joined the ends of all three wires (the "Y" point) and found they were soldered together. I got a different meter and tested it again... and everything seemed to be fine! Was it just that meter? The coil wires were soldered together and a piece of sleeving was slid over each bare join. Finally I started sliding these pieces of sleeving off the connections, and on just the second one, I found the wires had been tinned, but never soldered together. I went from wondering what could possibly be wrong to wondering how the motor could possibly have run for 5 years without ever an electrical problem until now.
   I also wonder if that loose connection may have had anything to do with my own motor controller's phase 'B' low MOSFETs repeatedly blowing up. Probably not, but it did keep blowing the same drivers.

   When I put it back on the car, for the first time I had meters that made it convenient to view both the DC current from the battery and the AC current to each motor phase at the same time. With the motor idling, about 1 amp of DC battery current (1a * 36v = 36w) led to about 6 amps of AC phase current in the first phase I checked. The trick is that the most of the AC current in a an idling motor is out of phase with the AC voltage and so it doesn't represent power actually being used. Or another way of looking at it for a BLDC motor is that most of the current energizing the coils is returned to the power line capacitors when those coils shut off. But this was the first time I could readily see the DC/AC currents difference in one of my motors. It also started to make somewhat more sense that my own motor controllers have trouble at higher currents: if 40 amps DC made 240 amps in the AC, that would be a lot more current being switched by the contoller. It probably isn't that high, but still it's probably substantially higher than the DC.

   But the phase currents weren't all the same, and the motor still didn't power through a complete cycle. There were still places it stalled or wouldn't start. The green light stayed on on the controller. It was probably 'off' for 1/3 of each turning cycle now instead of 2/3. On the 19th I took the motor apart again. I couldn't find anything wrong. That opened the possibility - seeming inevitability - that one set of outputs in the controller had died, notwithstanding that its trouble lights stayed green.
   There were two things that could be tried to be sure it was narrowed down: running the motor with another controller, and running a different motor with the Kelly controller in the car. Perhaps it was time to finally repair the Electric Caik Outboard, sitting now for over a year, and try running it with the Kelly controller? I decided on that course.

   The Caik's rotor was the 7.5" one that still didn't have the improved magnet bonding. I decided it should be upgraded. That way I would feel safe running it up to 3000 RPM instead of 2000. The boat could run at 24 volts instead of 16 or 18 and still go full throttle safely, for a faster boat ride. Assuming the controller didn't blow up (possible), the motor overheat (unlikely) or the 50 amp breaker blow.
   So on the afternoon of the 22nd I spent a couple of hours milling slots in the rotor so I could wrap the magnet strapping around the magnets and rotor, and another hour prying the steel filings off the magnets and then sanding them rough so the fresh epoxy would adhere. I just "eyeballed" the small slots, trusting they'd be close enough. The next day (23rd) I sanded the weak zinc coating (and the paint over it) off the back of the rotor with the belt sander to give the epoxy a good grip on the steel back. Then I cut the strapping, folded it in thirds and fed it through the slots (dry), and then epoxied it into place. When I had done the magnet side I flipped it over and did the back. I positioned the straps where they should be, put some thin PE sheet over each one, and put on some little steel weights that I have. I put an aluminum disk over the whole thing and put a heavy weight on top. I left it to set undisturbed, since everything would probably shift while moving it to the oven, and there was no rush.
Slots milled, and everything cleaned off and sanded          

Finished upgrade, except some of the epoxied fabric corners stuck out too far and
had to be snipped off with side cutters.

Checking the fit of the motor in the outboard.

   But several things then came up and I didn't get the Caik motor completely reassembled and back in the outboard by month's end.

   After taking the (Electric Hubcap) motor stator apart and reassembling it 3 times (not to mention probably previous times in years back), some of the 27 bolt holes holding the stator compartment together seemed to be stripped, the machine screws refusing to tighten. I really ought to stuff some epoxy and bits of polypropylene cloth into the holes, then re-drill the holes out to repair them... Or maybe drill them out larger and put in 1/4" nylon screws that would have no electromagnetic heating effects? I went to the drawer of plastic screws. There were just 9. The Hubcap motor would need 27. The Caik motor would need 18. But there was a box of longer ones that could be cut down. But this work too was halted until October.

Gear Cutters

   The set of gear cutters arrived on the 13th. They came with 27mm keyed holes in the middle. I scrounged a 1.25" x 2.5" piece of stainless steel shaft to make a center axle for them that would fit in the milling machine's largest collet chuck, and turned it to fit the same day. On the 15th I added a 1/4" machine screw hole to screw on a washer so the cutter couldn't fall off the end, and a shaft keyway so it couldn't slip instead of turning.

   Jumping ahead, I noticed the cutter had some wobble to it, and on the 21st I did some more work on it. I had to trim it down from 5/8" to 9/16" collet chuck size. I got it pretty true.
   The rotary table and dividing plates still hadn't arrived (and didn't during September), so I couldn't make an actual gear. I mounted a piece of plastic in the tool holder on the lathe and cut a few sample teeth. The plastic cut like butter, and my estimation of the practicality of making plastic gears went way up. OTOH, it had better not break or wear down as readily as it cuts! Someone in a video took apart a piece of equipment in which he thought its nylon gear would surely be worn down. But it still looked like new, and if anything, he said, the brass gear it meshed with looked slightly worn. This is encouraging!

Gear Making Calculations

   Later on the 15th I looked up how to do gear making calculations.

From the Website: http://khkgears.net/gear-knowledge/gear-technical-reference/calculation-gear-dimensions/
(From Japan. They still make things in Japan?)

Gear dimensions are determined in accordance with their specifications, such as Module (m), Number of teeth (z), Pressure angle (a), and Profile shift coefficient (x). This section introduces the dimension calculations for spur gears, helical gears, gear rack, bevel gears, screw gears, and worm gear pairs. Calculations of external dimensions (eg. Tip diameter) are necessary for processing the gear blanks. Tooth dimensions such as root diameter or tooth depth are considered when gear cutting.

4.1 Spur Gears

Spur Gears are the simplest type of gear. The calculations for spur gears are also simple and they are used as the basis for the calculations for other types of gears. This section introduces calculation methods of standard spur gears, profile shifted spur gears, and linear racks. The standard spur gear is a non-profile-shifted spur gear.

(1) Standard Spur Gear
Figure 4.1 shows the meshing of standard spur gears. The meshing of standard spur gears means the reference circles of two gears contact and roll with each other. The calculation formulas are in Table 4.1.

Fig. 4.1 The Meshing of Standard Spur Gears
( α = 20°, z1 = 12, z2 = 24, x1 = x2 = 0 )

Table 4.1 Calculations for Standard Spur Gears

NOTE 1 : The subscripts 1 and 2 of z1 and z2 denote pinion and gear

All calculated values in Table 4.1 are based upon given module m and number of teeth (z1 and z2). If instead, the module m, center distance a and speed ratio i are given, then the number of teeth, z1 and z2, would be calculated using the formulas as shown in Table 4.2.

Table 4.2 The Calculations for Number of Teeth
Note, that the number of teeth will probably not be integer values when using the formulas in Table 4.2. In this case, it will be necessary to resort to profile shifting or to employ helical gears to obtain as near a transmission ratio as possible.

(2) Profile Shifted Spur Gear (Huh? More than I want to know... I hope!)

   I decided that the next thing I should do would be to cut a pair of sample gears... but the rotary table and dividing plates hadn't arrived yet. Well, at least I could calculate them. The cutters were M4. I presume all the dimensions are in millimeters. How about a 16 and a 27 tooth gear pair, just for somewhere to start?

So per the illustration 4.1:
1. M = 4
2. alfa = 20°
3. z1 = 16 z2 = 27
4. center distance (16+27)*4/2=86mm
5. reference diameter d = zm: 16*4=64; 27*4=108 (wasn't that easy!)
6. base diameter = d*cos(alfa): 64*cos(20)=60.14; 108*cos(20)=101.4868 (also simple... with a calculator)
7. addendum 1*m: 4; 4
8. tooth depth h = 2.25*m: 9; 9
9. tip (outermost) diameter da = d+2*m = 64+8=72; 108+8=116mm
10. root diameter df = d-2.5*m = 64-10=54; 108-10=98.

4. z1 + z2 = 2a/m = 2*101.6/4 = 25.4 teeth. Hmm, that might not be very convenient. Let's make it 100mm and hence 25 teeth.
Apparently these teeth can be divided as desired between the 2 gears. But that's not very many teeth!

For 16 teeth, 360/16 is 22.5 degrees between teeth. One would use cutter #2 (of 8), for gears with 14 to 16 teeth.

For 27 teeth, 360/27 is 13.33333 degrees. Use cutter #5, gears 26 to 34 teeth.

Starting from a disk of diameter da, one would cut in each tooth until the inner diameter was down to df. The reference diameter of of the 27 tooth gear is over 4". A 60 tooth gear would be almost 9.5". Excellent! Gears with the large M4 teeth I chose to buy (it was just a rough guess) are big in diameter. That's exactly whet I wanted to make: big, fat plastic gears. They should handle the forces involved in moving cars by virtue of their size reducing the loading at all points. At least that's my plan! And UHMW gears will be as slippery as teflon (PTFE), so they should be very low friction with little to no heat buildup, and low noise.

So, that's 43 total teeth, so the reference diameter between axle centers with M4 teeth is 43*4=172mm. The speed ratio is 27/16=1.6875.

   A differential gear unit composed of spur gears only (below) solved a key problem of having to make either a planetary type gear with an inside ring gear, or a differential gear with beveled gears. It somewhat belatedly occurred to me that I can't fit the milling machine into the inside of the ring to cut inside teeth, and I expect beveled gears require more than just the simple cutters, too. Now only spur gears will be needed. I'm counting on the plastic material to make them as smooth and quiet as metal helical gears... or at least smooth and quiet enough.

For this, I chose a ready-made housing, to hold 50 and 47 tooth m4 main gears. The gear diameters are:
 50*m=200mm (~7.8"), and
 47 teeth, 188mm (7.4").

That should leave room for the configuration discussed below.

Other ways to Make Plastic Gears!

   I already knew the 3D printer was out. The materials available weren't strong, slick UHMW, and finished pieces weren't as strong as those made from any solid plastic. The surfaces were rough, and they would quickly fall apart if handling stresses.

   Well, I had bought the stuff, looked up the instructions, and did the calculations above. I followed all the 'regular' steps and was merely waiting on the rotary table and dividing plates. After all that, late in the evening (still the 15th) it occurred to me to look at youtube and search for plastic gear making.
   The first thing that occurred to me as I looked down the list of choices without even watching any of the videos was that those rather large teeth could doubtless be molded instead of milled. In a video - just the thumbnail image - the person was casting gears of epoxy(?) in rubber moulds. Instead of a rubber mold I could make one injection mold for each desired gear. (for a production model) Then they could be mass produced in UHMW. I would only have to make an inside-out gear for the mold. (Now there's a way to make an inside ring gear... it would be on an outside! But then what about all the other gears?) Injection moulding would allow for more complex shapes, such as a ring gear with gears inside and out.

   Next I went to a simple 78 second video titled simply "Making Gears Outta Plastic".  No voice, just music... and just about fell out of my chair! A big gear with 17 big teeth was being cut out of a fat piece of UHMW by a CNC router much like mine! What could be simpler to do, or (for me) require less investment in equipment? Forget the involute gear cutter set I'd just spent hours making a mandrel for, forget the rotary table and dividing plates. It could all be reduced to numbers in a spreadsheet or CAD program, and a 1/8"(?) router bit with a long shaft! Inside and outside gears could be cut!
   A 1/8" router would make for slow cutting, but it would be the largest router that wouldn't horribly deform "m4" gear teeth. They might still be a bit distorted right at the inner tips of the spaces between teeth.

   Now, if only I had started thinking about alternatives and watched those youtube videos before I ordered all the gear making stuff!

   Then I started thinking about how hard all that math would be for the shapes. What if I used OpenSCad to design the gears, designing only a single layer thick, and then ran Pronterface as if I was going to print it to the 3D printer (and maybe actually print, just to see it). I could then take the G-code file generated by Pronterface and... wait a minute... Pronterface wouldn't account for the diameter of the router bit. I should probably order the 800$ "Rhino Express" that Victoria Waterjet uses. Yikes! Canada Revenue hadn't come through with my refund for 2015 yet (it did arrive later) and not to mention credit card debt, my bitcoins were disappearing rapidly. I wished I'd bought that instead of all these metal gear making tools.

Spur Gear Differential Gear & Variants

   On the 17th I searched for differential gears on the web - thinking perhaps to buy - and came across (wouldn't you know it?) Wikipedia. Once again I was surprised by finding something I had had no inkling existed: the first image in the article showed a differential gear made entirely of spur gears - no bevel gears, no inside ring gear. (Beautiful pencil sketch!) Thus it would be much easier to make if one is making the gears. Or one might buy the simple gears that make it up. If one wants the "no lube" effect of plastic, big steel gears could be purchased and the small ones done in plastic. In the drawing, it looks like one could fit 6 or 7 pairs of planetaries around the perimeter to spread the loads out.
   It's sort of a variation of what I called the "double sun planetary" I had conceived a few months ago. (I already suspected the gear arrangement I "invented" could hardly be very original in a well explored field like gears!) Doing the planetaries in counter-rotating pairs to make two equal ends at 1 to 1 ratio, and having all three elements rotating the same way, is one more demonstration that there are mechanical geniuses out there. (But it's my differential variable transmission idea, arf! arf!)

A spur gear differential constructed by engaging the planet gears of two co-axial epicyclic gear trains.
The casing is the carrier for this planetary gear train.
(The planetaries turn against each other in pairs as well as against the sun gears.
That's why the teeth extend into the middle. It took me a bit to figure this out.)

   The fixed drive would turn a gear or sprocket on the outside of the casing. Now, we don't need symmetry for the differential transmission. If one side had a larger sun gear than the other, its planetaries would turn outside the limit of the other gear, and their teeth could go right across. The other sun gear would have narrow "planetaries" connecting it to the outer ones. The sun gears could then be adjacent with just a tiny gap, narrowing the profile. It would also be a differential with a different ratio between the two sides. Probably one end could be given the reduction ratio wanted for the variable drive. But the larger the difference between the two sun gears, the larger the planets of the smaller one would have to be for their axles to clear the larger one - unless their axles didn't go across to that larger gear.
   Hmm! Maybe my own 'double sun' version would be better for the transmission? There's one less set of meshing gears, and the shell rotates the opposite direction, like a planetary gear. Later I realized that was what the choice was: if one drive turned in the opposite direction to the other (eg, one was driven through a pair of gears, causing its direction to reverse) one would want my configuration, but if both drives went the same way, the modified differential would be the choice.
   But whatever the exact gear configuration inside, I like the case layout. Somehow I was still wondering how to mount my gears of this form, and this is the now obvious solution.

Configuring a Production Model

   From the various bits, pieces and ideas, by the 20th I started considering how a production model might actually be configured, and "mocked up" the layout with the various potential components or substitutes for them on a piece of 11" * 17" paper. There are four main components within the case:

 * The motor shaft
 * The main differential/planetary/epicyclic gear (hereinafter "the differential". The "slow speed" end is the output shaft.)
 * The fixed drive from the motor shaft to the differential center/case (gear, chain or belt)
 * The variable drive from the motor shaft to the "high speed" end of the differential (variable pulley belt plus gear, chain or belt reduction - or - variable pulley belt with internal reduction in the differential unit). A control lever sticks out of the case to control the pulley and hence the final drive ratio.

   If the differential has the reduction built into it, there's no need for a chain drive reduction. And it would be undesirable since it needs lubrication. Why then would there be a third shaft? The next morning I had the potential answer: an idler pulley on a stiff arm could set the ratio, being variably inserted to force the variable pulley to smaller diameters. But on checking it out with the actual pulleys and belts, I found that the difference in belt length between fully out and fully in was so great that an idler would have to push the belt in so far it would hit the other side of the belt and more. An idler pushing outward might work in theory, but mostly it would only serve to reduce the grip of the belt on the pulleys and probably make it slip. And like a third shaft, it would create a "knee" forming a triangle which would make the whole thing bulkier.
   Reducing to one variable pulley for a production model started to sound rather tricky after all. Maybe a longer belt to make room for two idler pulleys for the belt to 'wrap around' - one in, one out? Those could be on a single pivot pin, which would pivot on a common center, twisting the belt between the two idlers to pull it in and change the variable ratio. Hmm, that would twist the belt too much.
   I decided the third shaft would be needed after all. But instead of a chain, I would prefer to use plastic gears if possible. When I layed out the components I found that if I made the gear drive go to the motor shaft and the belt drive go to the differential, the pulleys wouldn't stick out farther than the 10" differential required anyway, and so the shape could be rectangular. But it proved more practical to have the reduction external to the differential, so that wouldn't work. The bulge would have to stay... unless a short variable belt with just the right size of pulleys could be fit in.

   Perhaps interestingly the differential could pretty easily be made as a sealed oil bath type unit with metal gears if desired. I thought of the Raybestos brake drum I had used long ago in an attempt to make a centrifugal torque converter. It had one solid steel side and a thick solid steel rim. I found this hacked-up part at the bottom of the shelves and pulled it out. (What a junk collector I am! I was unable to find a second one I thought I had - must have tossed it?!?) If the inner rim was turned off on the lathe, a round disk of steel plate could form the other side and be bolted securely into exact place for the center bearings and planetary axles to go straight across. It was just big enough to hold some good size gears about 5/8" wide. I would have liked wider. It might need those metal gears. It seemed awfully heavy, but so was the casing on the original planetary gear on the car. Given the torque and radial forces, this case needs to be very stiff. So this unit should be excellent for a prototype - perhaps even for a production model. (The Raybestos model/part number is in some ancient issue of TE News or old motor making manual.)
   Or, I have a couple of pieces of aluminum pipe, 9" and 10" diameter and 3" or 4" long. And at least the 9" can be bought readily. With aluminum, turned to fit (and maybe one side welded on), they'd hold wider plastic gears. And they'd be lightweight. Then again, steel is stronger... I should stick with it. And the brake drum is already precision turned. If I make the precision planetaries axle holes by CNC waterjet in the plate for the second side, I can use it as a template to drill out the ones in the drum.

   Another thought is that the cast Raybestos drum is 10" O.D, 1.75" wide, flat on the outside and has a rough finish. Is this not ideal to be turned by a fat 99% efficient flat belt instead of either a gear or a chain? No extra width for that! (Or if that slips, the rim is thick enough to cut toothed belt teeth into.) In fact, 2 of the 5-V poly-V belts I already have one of, driven by 2 poly-V pulleys, might be about right. On the 30th I bought a 2.75" poly-V pulley for it, giving a reduction of about 3.5 to the drum - not dissimilar to the 3.0 of the chain on the Sprint. It starts to look doable!

   There'd have to be some serious calculations to have all the gear sizes, case size, belt lengths and overall chassis size and shape come together. Maybe mock-ups, incorporating some paper or cardboard, as a reality check on them. But perhaps this whole 'production model' thing might start to come together in a relatively doable form after all?

2nd "Mock-Up", with variable belt to differential unit & fixed drive to input shaft.
   I started a "mock-up" on the 20th with an overall chassis layout using all the parts I had tentatively determined on. Allowing for a 'Raybestos' 2.5" wide differential/epicyclic gear, bearings and the variable pulley, I came up with about 5.5" thick, by 13" wide by 16" long - inside dimensions. The width was allowing for a third shaft with a 6" pulley and a gear pair (one gear represented by the off-white circle) driving the end of the epicyclic gear. But really that width only needed to be a bump in the side of an 11" wide space. And of course measures like rounding the ends could further reduce the overall profile.
   If the 'flat' belt is a poly-V belt like the one shown (or two side by side for more strength and better grip), the small pulley(s) could easily be a genuine poly-V pulley, which would keep the belt in line on the large flat surface. (I don't envy the person who has to change the flat belt. One side of the unit will have to come right off.)
   Soon I noticed that if the gears went to the motor shaft and the variable belt part to the differential shaft, instead of the other way around, it could be reduced to 11" wide (needed for a 10" epicyclic gear regardless) because the motor shaft was offset from the center, leaving more room. And then with a slightly longer belt than I had, the protrusion (gear attachment permitting) could be pointed inward to chop an inch off the thickness to 6": 16" x 11" x 6". If not, perhaps the protrusion could just be a bulging arc cover?

   I started thinking about the many innovations, that there were so many of them in one project. Too many? Anything that didn't work could set things back considerably. The first thing was making plastic spur gears for the differential. Should I just buy some metal gears for the first tests? I thought about that until the 21st, when I found that cutting plastic gear teeth was simple - the UHMW plastic cut like butter. (except for a bunch of shavings to trim off after) So I decided to do it the other way around: go with plastic gears unless they definitely didn't work. A 6" diameter gear 1.25" wide, with 6 planetaries compared to a steel sun gear from a Chrysler, would be, um:

diameter: 45/152.4 = .295
width: 15/31.75 = .472
planetaries: 4/6 = .667

Total relative loading: .295*.472*.667=.0929, or 9% as much force on the large gear teeth surfaces as on the small commercial ones. That sounds promising in itself. However, it has to be tempered by the fact that it's our output gear, whereas in the Chrysler it still goes (at the least) through some final reduction gear before the wheels. The tentative conclusion is that unless there's a final reduction to the car wheels, the gear might not be strong enough. One could of course incorporate the 3 to 1 chain reduction in the Sprint now following the variable transmission (or something similar in another vehicle) to achieve that end, perhaps running 1 to 1 (minimum reduction) through the variable transmission part. That would allow driving both wheels again.

   A better comparison might be with the gears out of the Sprint manual transmission. All five gears plus reverse run on the same shafts, with only one meshing connection between each input and output gear, and they're followed by a 4 to 1 final reduction to the differential gears before the wheels. I didn't get these out to check by the end of the month, but they seem to take a lot of torque for a small size.

   With the 'brake drum' body for the differential the gears could only be about 15mm wide. OTOH the large gear could be up to almost 8". These would modify the force per gear tooth area calculations somewhat. On the 25th, I thought of expanding the whole thing to use a 3" wide * 12" diameter differential, allowing very large and wide plastic gears. But I went back to the solid, already made and machined Raybestos drum.
   Considering the differential gear element spacings, I decided I would make an 8", 50 tooth large-end sun gear, leaving just enough radius outside for the planetaries, and a 47 tooth small end. I would have liked to go much smaller for an internal gear reduction from the variable belt driven end, but by making it just a bit smaller than the other, the axles for its planetaries could go right across without hitting the large one, simplifying construction. The speed reduction can almost as easily be external.

   I reluctantly concluded that to do that, at least for the prototype, I should use a chain drive. A gear drive would reverse the direction, and I decided that the unidirectional differential would be easiest to make. (For a production model, one might use three gears instead of two to double reverse the direction, and the mid gear might also take it sideways, which could shave a layer off the overall thickness, potentially bringing it down to around 4".) With the last mock-up on the 30th I decided to add a inch to the width, and it looked like I could take one off the thickness and make it rectangular, 12" * 16" * 5" - inside dimensions. Maybe with the bulge for the variable pulley.
   I couldn't make gears until the rotary table and dividing plates arrived. But I had them pretty much sized up. It seemed the next things to do were to were to find roughly 3.3" pulleys for the poly-V belts, and when I had all the dimensions to start laying out the steel plate parts for the other side of the drum and the outer case.

   From about the 25th I started hunting for a new computer to run "Rhinoceros 3D" CAD program for designing these metal parts, to be cut by CNC waterjet. (also to run "Sibelius 8" music notation program for composing orchestral & concert band music.) Newer versions of those won't run on my old Mac and of course they don't offer any of the older versions that did and still would if I could get them. And there are no Linux versions. (The one I wrote my music with, "Overture 2", which was a great program, has been unsupported since USB was created, and doesn't run on any newer computer or O/S.) A knowledgeable friend said Windows software can be run under Linux and had no reservations about being able to get them going. So I got a new (used) box that's actually running Linux. But it has much more memory and CPU power than my old Linux laptops.
   While we were setting it up we found "MuseScore" and installed that. IF I find when I have time that it'll do the job I can skip buying Sibelius. And "Blender 3D" was mentioned. IF that'll do the job I may pass on Rhinoceros too. Those would be big savings over the commercial software even if I donate pretty nicely to the creators - as was buying a 'government surplus' used computer.

   As I wrote this, I thought again of "OpenSCad" that I already use for the 3D printer. When you're starting with numbers, positions and sizes, it's quite easy to just enter numbers and hit "render" to see the result. No learning curve. OpenSCad is after all, open source, and theoretically I could make a version, or a ".DXF" file exporter, that (as required by Victoria Waterjet) defines curves (as you originally define them) instead of converting them into a series of short straight line segments. But the learning curve for doing that programming would surely be beyond anything I'd want to attempt. And it wouldn't help for doing CNC router porjects where the diameter of the router bit needs to be taken into account.

Electricity Generation

Conversion of Car Alternator to Permanent Magnet Alternator

   I started this thinking I'd want an alternator(s) for testing wave power units. The wave power has been put aside, but having started on the alternator, and it showing the promise of being a small project that could be completed quickly, I thought I would finish the conversion. Then I might try it out, perhaps as a small hydro generator with a primitive pelton wheel, just with the garden hose. (Maybe in a windplant? Well, later!)

   The trouble with car alternators as alternators/generators for windplants and the like is that they have a 'field coil' electromagnet in the armature, which is what the brushes power. This uses power. The more the desired voltage per RPM, the more current has to go into the field coil to give it more magnetism. In one alternator I measured 3 amps at 12 volts, or 36 watts, being consumed by this coil. That's probably the maximum, but if you don't use that, you need some way to give it some lesser voltage. And then it will need higher RPM to get the same output. If your small windplant is making 50 watts in light wind at a lower RPM, consuming 36 watts would only leave 14 watts output.
   A permanent magnet on the other hand uses no electricity, so the whole 50 watts is available. If a 'supermagnet' (eg an FeNdB 'neodymium' magnet) is used, the flux should be very high, giving maximum power per RPM. The disadvantage is that its voltage (and the current capacity) rises linearly with the RPM. Typically it won't work at all until the RPM is high enough to give voltage higher than batteries being charged, and above that it will put more and more current into the battery regardless of its state of charge. I put a very powerful ring magnet in in place of the coil. I thought it would probably give more volts per RPM than the coil at full strength.
   There are other ways to regulate output voltage (once rectified to DC) for battery charging or other load. A DC to DC converter with a fairly wide input range will go a long way toward it. And maybe when it's done, the infinitely variable transmission (probably under computer control) can have it maintain some desired RPM and hence output voltage level regardless of propeller (or other input) speed - and within limits of available input power, regardless of varying output loads.

   Many conversions are done by replacing the entire armature, with one having supermagnets spaced around the outside facing the coils in the typical 3 coils per 2 magnet poles 3-phase configuration. This technique also works for most any three phase induction motor. The biggest problems are making the new armature, making it solid enough that it won't fly apart at higher RPM.s, and that ideally the magnet outer faces should be curved to match the diameter to get close magnetic coupling.

   My plan for this conversion was a little different: Replace the toroidal electromagnet coil in the armature with a toroidal (ring, donut) shaped supermagnet. The rotor's soft magnetic 'petals' would be the alternating north-south 'facing magnets' as before, carrying the magnetic field of the magnet in the center. Since it was unchanged, they would be the right number of magnets, spacing and flux gap to the stator. It sounded really simple, but the details concealed devils. The first one was getting the old thing apart, which wasn't trivial as described last issue. The second one was to fit the magnet into the unit when it had a different inside diameter than the coil.

   On the 5th I went to turn down the 'coil center cylinder' pieces that were inside the coil. They were too large in diameter for the magnet with its smaller center hole, and I figured they'd short out its magnetic field. (hmm, I think I have that wrong?) So I was going to turn away most of the center iron from inside the coil on the lathe, just leaving a 1" diameter by 1/8" round 'button' on each end to mount the magnet on.
   But owing to various protrusions there didn't seem to be any way to attach the pieces to my small lathe. I had to go down to AGO the next day and get machinist Ralph to do it for me on a very large lathe with a big chuck that could grab it around the outside without bending the fan blades. I had hoped this could be a home project for any handyman. It was certainly not in my thoughts that I'd have to have outside help myself! (The mid-size lathe at Makerspace might have worked, but I wasn't at all sure it was big enough, and I didn't want to drive all the way out there only to find it wouldn't. I had already spent 2 hours in surprisingly heavy midday traffic to get to just a couple of places, once again double the time it should have taken.)
Pressing bearing back into housing.
   On the 7th I pressed the armature back onto the shaft. The first time I didn't quite get it to the end, and the case wouldn't fit back on. The second time I put too much pressure on it and the needle moved on the gauge. (A ton?) Then I remembered it was the ball bearing I was pressing on. It dented the race or the balls and now it runs rough.
   But that wasn't the only problem closing the case. The coil had been 24mm wide and the magnet was 25.4. Furthermore, Ralph couldn't quite get to the bottom, so there was an extra millimeter of thickness at each end. I could see that the fan blades hit the end before it closed together. I opened it again and took out a 6mm thick washer I was now sure would prevent it from closing, and I bent the fan blades. In doing that they swung outward a bit and wouldn't go past the stator. I snipped the protrusions off. It still didn't turn and I hammered the blades down some more. Now they were well recessed (and would move ~1/4 as much air) but it still didn't turn. Then I realized some of the fan blades on the other end were bent and hitting the connection wires where they passed by. I stuck in a screwdriver and bent them straighter again.

   When I had it together there seemed to be too much friction. I tried more than once to take it apart and redo it, to no avail. All I accomplished was to break a minor crack in the housing. I think it's magnetic friction, in spite of there being no apparent cogging. (Unless I did actually bend the shaft a bit when I took it apart. But it seems too uniform all the way around for that.) I put the poly-V belt pulley back on the shaft, and with the extra grip of the pulley instead of just the shaft, it was easier to spin and didn't feel absurdly stiff. It still came to a stop within about half a second when spun. The next morning it seemed better all by itself - anyway better than the lawnmower motor, my other sample of a smallish permanent magnet motor. Were my expectations just too high? I guess the magnetism is bound to make for some friction compared to something without magnets in it. The alternator shaft attracts steel pieces. Not quite satisfied yet, I removed the brushes from the lawnmower motor. Now the drag was comparable. Must be an inevitable magnetic effect. Freely as it spun before the conversion, it doubtless would have had a similar drag effect running with the electromagnet on, so really no efficiency has been lost in the conversion. However I do believe it's possible to make permanent magnet motors and generators that don't have this drag by careful design. It seems to me I've seen some very large ones that turn considerably more freely.

Spinning it

   14 volts DC at 20 amps continuous when rectified seems reasonable for a small car alternator. That would be 280 watts or 1/3 HP. You wouldn't turn that power by hand, and not for long with bicycle pedals! With a 24/7 driving source, that would be a lot more, more reliable overall power than a 260 watt solar panel.

  By giving it a spin it by hand, with the oscilloscope connected between two phases, it produced about 12 volts peak to peak at about 50 Hz.

Project Creep!

   Effectively the project was finished: I had converted the alternator. Of course, I had no idea what RPM it got up to or how it might perform. It needed to be connected to a motor and spun at a steady speed. Uh-oh! This was supposed to be a short, simple project! But what was the point to converting it without testing it to see how good it was?

   In order to do that, on the 8th I dug out a poly-V belt pulley from the Sprint engine parts that matched the alternator's pulley, and cleaned the thick grease off it. (yuk!) Then I did likewise, as best I could, to the belt. (And I wish I had cleaned the alternator better when I had it apart!) The pulley was the water pump pulley, and it had no center. I thought, what better center than the shaft and end off the water pump? I went out to the engine and removed the water pump. That started a whole comedy of events. I couldn't get it apart. Trying to press the shaft out, the case cracked and half of it flew out of the press. (And here I had neglected to put on safety glasses!) Then I figured if it was that flimsy, I would hammer it apart on a rock with a big maul. With safety glasses on, I took a swing. Nothing broke. Instead the pump flew straight at my shin, leaving a good bruise. I gave up for the day. The next day (9th) I started cutting the 'pot metal' case apart with the angle grinder. With a slit down each side it finally snapped apart when hit with a hammer onto a screwdriver in the biggest slit. Even then, the shaft still had two big bearings covering its whole length and seemingly no possible way to press them out. I finally had to cut these apart as well, and then I finally had my shaft with attached pulley mounting. I think it would have been easier to make one from scratch.
   I found 4 bolts that fit it in the Sprint bolts box and put it together.

Converted Alternator Performance

 The shaft was too big to fit in the drill press chuck, but it fit in the lathe chuck. I set the lathe to medium speed, put the belt on, and simply held the alternator in the correct position with the belt in line and relatively taut. Calculating from motor speed and pulley sizes of 3 sets of pulleys, I got the following result between two phases: 18.5 VAC (RMS), 228 Hz, at a calculated 2435 RPM. The frequency doubtless indicates it was actually 2280 RPM. Shorting two wires together (through a light alligator clip test lead) gave about 18-19 amps AC. Presumably shorting all the outputs would have yielded more. (I forget the math - 19 * sqr.root of 2 = 27 Amps AC? or was it sqr.root of 3?) That probably means you could get the 20 amps at 14 volts DC after it was rectified - at 2280 RPM. Higher RPM would give more potential, but at some point it will overheat. The poly-V belt pulley sizes raised it from 1068 RPM to 2280 with no gear noise and probably quite low losses, so that would be the way to go with a lower RPM energy source, rather than trying to find a lower RPM alternator.
   I reduced the lathe speed and got 10.4 VAC at 130 Hz/1300 RPM. 10.4 * 2280/1300 = 18.24 VAC, the voltage measured at 2280 RPM, demonstrating that voltage is indeed linear with RPM. (It's probably even closer than the calculation - the RPM.s aren't exact... from the meter reading jumping between 228 and 229, and 129 and 130, I estimate they were probably closer to 2282 RPM and 1296 RPM. And the voltages only had one decimal point on this meter too.)

   It seems in line with the performance I'd expect from this small alternator, but without wasting maybe 20 or 30 of whatever watts were generated to activate the field coil. Thus, it seems like a good way to convert a car alternator to permanent magnet. It probably took too long and cost too much for a one-off unit that I have no particular use for with the wave power project set aside.

   Other options would be of course a 120V, 12A lawnmower motor, which would need no conversion, or perhaps to build the frictionless Scorag/Piggott axial flux windplant's alternator from scratch or from available kit components. Those run at good lower RPM.s for a typical small windplant. Or maybe there's good PM alternators or generators for sale at good prices these days, perhaps from China?

  Is it a way someone might earn cash?

   I have the thought that someone might get good at taking alternators apart and fitting them with ring magnets, then selling them to people who want to make energy generating devices. If one had it down pat, and found which models converted the most easily, it shouldn't take so long to do one. It would be simpler and cheaper than converting induction motors. To do it, one would need a press, a lathe, a source of cheap alternators, and to buy a supply of ring supermagnets. Don't try to convert a very small alternator like the Sprint's with a coil thinner than the magnet. Expanding the armature length just makes for much fiddling around to get things to fit. If the coil is thicker than the supermagnet, leave some extra metal when turning down the inside so that the armature size doesn't change.
   But I would see it as a means to an end, that it might open up further opportunities. For more value added, make whole energy generator units like VAWT.s or pelton wheels and sell the whole thing? (Aren't there any good small VAWT.s for sale commercially - just buy and hook up? There must be a market! Days later: gosh, see below!)

Tesla Turbine Windplant?

   After considering using a Tesla Turbine for the OWC wave power's air turbine, and having at least for now dropped the wave power again, and having made a spare permanent magnet alternator (above), an e-mail said the sample Tesla turbine I'd ordered had at last been shipped. On the 18th I started thinking about using one for wind power. They seemed to be sort of like having a multi-stage turbine all in one unit, which might actually be able to exceed the theoretical "Betz limit" of about 58% efficiency for a single stage unit such as a horizontal propeller, to get more power per square meter of wind frontage.
   Perhaps one could make a "horn" and aim it into the wind to focus air pressure into the intake. If Teslas were really so efficient when used at lower air pressures, might it not be better than anything else, or at least better than a VAWT? A big rectangular "horn" pivoting from the front would always aim itself into the wind. It could be made out of plywood or other handy material, and have a square or rectangular profile. It could have spring panels on the sides, so that if the wind was too strong, they would open to prevent damage both to the turbine and to prevent bending of the axle. Being horizontal axis it would need a slip ring to prevent twisting of the output wires. But there would be no spinning parts on the outside for birds or anything else to run into.

   I decided I could make a very miniature version to try out with the tiny test model. Perhaps at this scale one could make various shapes and sizes of horns from paper and tape to see which shape worked best, or if it mattered very much. And at that scale if there was no wind, there was always a fan or the vacuum cleaner on "blow". Later that day, some sensors that had been intended for the wave power arrived. Some of them would be useful: an anemometer and pressure sensors to record performance data, either by microcontroller or just by hand. Ideally I'll want to know the air velocity and pressure inside the "horn" at the turbine intake compared to that downwind at the output. But the turbine still hadn't arrived by month's end.

Victoria BC Canada